Re: [codec] USAC as reference?

[David Virette <[e-mail]>]

Hi Jean-Marc,
Of course IETF can liaise with MPEG asking to include OPUS in USAC
verification tests, but I am a bit skeptical that OPUS will be included in
their extensive testing if a sufficient testing effort is not done in IETF
before.
By extensive listening tests, I mean a verification test where more than 15
listening labs will take part to the first round of listening to conduct
between 1 and 3 experiments each.
Best regards,
David


-----Original Message-----
From: Jean-Marc Valin [mailto:[e-mail]] 
Sent: vendredi 15 avril 2011 17:50
To: David Virette
Cc: 'Christian Hoene'; 'Anisse Taleb'; [e-mail]
Subject: Re: [codec] USAC as reference?

Hi David,

On 11-04-15 10:02 AM, David Virette wrote:
> The only reference quality implementation of USAC is owned by the initial
> proponents of the codec. If the group wants to test OPUS vs USAC, the best
> way is probably to inform MPEG that IETF would like to conduct some tests
> comparing those two codecs. Regarding the inclusion of OPUS in the USAC
> verification tests, the test plan and accompanying workplan have already
> been agreed and the first round of testing will occur in June. However, it
> is foreseen to have a second round of testing by the end of the year. It
> could be proposed to include OPUS in this second phase. As Anisse said,
> there will be an MPEG meeting in July, this leaves enough time to liaise
> with MPEG before the next meeting. But I think that only a standardized
> codec could be included in the USAC verification tests.

I think it's indeed a good idea to liaise with MPEG before the next meeting 
to get Opus included in the second round of USAC testing. I think the 
condition that only standardized codecs can be included again supports the 
idea of first publishing Opus as a proposed standard to facilitate further 
testing.

Cheers,

	Jean-Marc